Update app.py

app.py

@@ -1,103 +1,86 @@
 import cv2
 import numpy as np
 import matplotlib.pyplot as plt
+import streamlit as st
 from skimage import measure
+from io import BytesIO
 
-def preprocess_image(image_path):
-    # Read image
-    img = cv2.imread(image_path)
+def preprocess_image(image):
+    img = cv2.imdecode(np.frombuffer(image.read(), np.uint8), cv2.IMREAD_COLOR)
     
-    # Check if the image is loaded correctly
     if img is None:
-        print(f"Error: Unable to load image at {image_path}")
+        st.error("Error: Unable to load image")
         return None, None
     
-    # Convert to grayscale
     gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
-    
-    # Apply Gaussian blur
     blurred = cv2.GaussianBlur(gray, (5, 5), 0)
     
     return img, blurred
 
 def detect_edges(blurred_image):
-    # Detect edges using Canny edge detector
     edges = cv2.Canny(blurred_image, 50, 150)
     return edges
 
 def find_contours(edges_image):
-    # Find contours in the edges image
     contours, _ = cv2.findContours(edges_image, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
     return contours
 
 def filter_contours(contours, min_area=1000):
-    # Filter out small contours (e.g., noise, unwanted areas)
     filtered_contours = [contour for contour in contours if cv2.contourArea(contour) > min_area]
     return filtered_contours
 
 def draw_contours(image, contours):
-    # Draw contours on the original image
     output_image = image.copy()
     cv2.drawContours(output_image, contours, -1, (0, 255, 0), 3)
     return output_image
 
 def calculate_area(contours):
-    # Calculate the area of the detected contours
     areas = [cv2.contourArea(contour) for contour in contours]
     total_area = sum(areas)
     return total_area
 
 def remove_shadows(image):
-    # Convert to HSV to better detect shadows
     hsv = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
-    
-    # Define shadow color range in HSV
     lower_shadow = np.array([0, 0, 0])
     upper_shadow = np.array([180, 255, 80])
-    
-    # Mask for shadows
     shadow_mask = cv2.inRange(hsv, lower_shadow, upper_shadow)
-    
-    # Remove shadow by inpainting (filling shadow areas with nearby pixel values)
     result = cv2.inpaint(image, shadow_mask, 3, cv2.INPAINT_TELEA)
-    
     return result
 
 def main():
-    # Ask the user to provide the image path
-    image_path = input("Please provide the path to the rooftop image (e.g., 'image.jpg'): ")
-    
-    # Step 1: Preprocess image
-    img, blurred = preprocess_image(image_path)
-    
-    # If the image was not loaded successfully, exit the program
-    if img is None:
-        return
-    
-    # Step 2: Detect edges
-    edges = detect_edges(blurred)
-    
-    # Step 3: Find contours
-    contours = find_contours(edges)
-    
-    # Step 4: Filter contours to exclude unwanted areas
-    filtered_contours = filter_contours(contours)
-    
-    # Step 5: Remove shadows from the image
-    image_no_shadows = remove_shadows(img)
-    
-    # Step 6: Draw the predicted rooftop boundaries
-    output_img = draw_contours(image_no_shadows, filtered_contours)
-    
-    # Step 7: Calculate the area of the rooftop
-    total_area = calculate_area(filtered_contours)
-    print(f"Total Rooftop Area: {total_area} pixels")
-
-    # Step 8: Display the result
-    plt.figure(figsize=(10, 10))
-    plt.imshow(cv2.cvtColor(output_img, cv2.COLOR_BGR2RGB))
-    plt.axis('off')
-    plt.show()
+    st.title("Rooftop Area Calculation from Satellite Image")
+
+    # Allow the user to upload a file
+    uploaded_file = st.file_uploader("Upload a rooftop image", type=["jpg", "png"])
+
+    if uploaded_file is not None:
+        # Step 1: Preprocess image
+        img, blurred = preprocess_image(uploaded_file)
+        
+        if img is None:
+            return
+
+        # Step 2: Detect edges
+        edges = detect_edges(blurred)
+
+        # Step 3: Find contours
+        contours = find_contours(edges)
+
+        # Step 4: Filter contours to exclude unwanted areas
+        filtered_contours = filter_contours(contours)
+
+        # Step 5: Remove shadows from the image
+        image_no_shadows = remove_shadows(img)
+
+        # Step 6: Draw the predicted rooftop boundaries
+        output_img = draw_contours(image_no_shadows, filtered_contours)
+
+        # Step 7: Calculate the area of the rooftop
+        total_area = calculate_area(filtered_contours)
+
+        # Step 8: Display the result
+        st.image(output_img, channels="BGR", caption="Detected Rooftop with Boundaries")
+        st.write(f"Total Rooftop Area: {total_area} pixels")
 
 if __name__ == "__main__":
     main()